Multiple works have applied deep learning to fringe projection profilometry (FPP) in recent years. However, to obtain a large amount of data from actual systems for training is still a tricky problem, and moreover, the network design and optimization is still worth exploring. In this paper, we introduce graphic software to build virtual FPP systems in order to generate the desired datasets conveniently and simply. The way of constructing a virtual FPP system is described in detail firstly, and then some key factors to set the virtual FPP system much closer to reality are analyzed. With the aim of accurately estimating the depth image from only one fringe image, we also design a new loss function to enhance the overall quality and detailed information is restored. And two representative networks, U-Net and pix2pix, are compared in multiple aspects. The real experiments prove the good accuracy and generalization of the network trained by the diverse data from our virtual systems and the designed loss, providing a good guidance for real applications of deep learning methods.

Accurate vector extraction from design drawings is required first to automatically create 3D models from pixel-level engineering design drawings. However, this task faces the challenges of complicated design shapes as well as cumbersome and cluttered annotations on drawings, which interfere with the vector extraction heavily. In this article, the transmission tower containing the most complex structure is taken as the research object, and a semantic segmentation network is constructed to first segment the shape masks from the pixel-level drawings. Preprocessing and postprocessing are also proposed to ensure the stability and accuracy of the shape mask segmentation. Then, based on the obtained shape masks, a vector extraction network guided by heatmaps is designed to extract structural vectors by fusing the features from node heatmap and skeleton heatmap, respectively. Compared with the state-of-the-art methods, experiment results illustrate that the proposed semantic segmentation method can effectively eliminate the interference of many elements on drawings to segment the shape masks effectively, meanwhile, the model trained by the proposed vector extraction network can accurately extract the vectors such as nodes and line connections, avoiding redundant vector detection. The proposed method lays a solid foundation for automatic 3D model reconstruction and contributes to technological advancements in relevant fields.

With the rapid advancement of three-dimensional (3D) imaging technologies, 3D data has been increasingly utilized in areas such as medical diagnostics, digital heritage preservation, and virtual reality, where protecting sensitive spatial information is essential. To address this need, 3D information hiding aims to securely embed depth data into a host image without introducing perceptible visual distortions. Unlike conventional two-dimensional (2D) information hiding, which focuses on pixel-level planar data, 3D hiding must preserve spatial structure, geometric consistency, and depth fidelity—posing greater challenges in terms of capacity, imperceptibility, and robustness. In this work, we propose, to our knowledge, a novel 3D information hiding method that integrates single-pixel imaging with diffusion models. A dual-stage encryption framework is employed to enhance both confidentiality and imperceptibility. Additionally, we introduce an adaptive weight allocation strategy that prioritizes regions of interest, thereby improving both hiding performance and recovery accuracy. Experimental results demonstrate that the proposed method significantly outperforms existing approaches in terms of hiding and recovery performance, showing strong potential for practical deployment in secure 3D data transmission and storage.

The 3D modeling of transmission towers currently depends on manual operations, resulting in high labor and time costs. To this end, an automatic 3D modeling technique based on 2D drawings is proposed. Using this method, the 2D drawings of transmission towers were analyzed first, then a 3D model of a tower was reconstructed using a counter-to-detail strategy. The analysis of the 2D drawings aimed to segment the geometric shapes and subsequently extract the vectors. All obtained vectors were classified into outer contour vectors and internal structure vectors. For each tower section, the 3D outer contour framework was constructed first using the wireframe model algorithm, followed by the assembly of internal details onto the 3D contour framework to fully reconstruct the 3D model. Experiments demonstrated that constructed 3D models exhibited high accuracy, with an average chamfer distance to the real scanned dense LiDAR point clouds of less than 0.05 m, which was less than 1% relative to the whole size of the created models. Furthermore, the automation of this technique implies its potential for various applications.